Skip to content

Guides

These guides show how to process spatial transcriptomics data with run_together, the recommended way to run CartLoader: one command per run that takes raw platform output to deployable CartoScope assets.

New to CartLoader? Start with Get Started. For every option of every command, see the Reference.

Question

run_together Guides

  • Single sample
    Single Sample: one command from raw output to CartoScope assets

    Read

  • Multiple samples
    Multiple Samples: one shared model across samples

    Read

Platform Tutorials

These tutorials process a small, representative dataset from each platform. Each one starts with the single run_together command for that data, then walks through the same analysis step by step with the individual commands.

Tip

For consistency, we use mouse hippocampus when available. Each tutorial begins with data in the platform's native format.

Imaging-based Platforms

  • Xenium
    10x Xenium Tutorial

    Read

  • MERSCOPE
    Vizgen MERSCOPE Tutorial

    Read

  • CosMx
    CosMx SMI Tutorial

    Read

Sequencing-based Platforms

  • Visium HD
    10x Visium HD Tutorial

    Read

  • Stereo-seq
    Stereo-seq Tutorial

    Read

  • Seq-Scope
    Seq-Scope Tutorial

    Read

  • Pixel-seq
    Pixel-seq Tutorial

    Read

Any Other Platform

  • Generic
    Generic Tutorial: any transcript table

    Read

For the exact inputs run_together expects from each platform, see Platforms.

Beyond the Guides

  • Step by step: run each stage (sge_convert, run_ficture2, run_cartload2, ...) yourself, for custom workflows. See Step-by-step.
  • Legacy orchestrators: the older run_xenium and run_visiumhd pipelines and their end-to-end walkthroughs are kept under Legacy Orchestrators; new work should use run_together.